Abstract
Accurate lesion metabolic response estimation is imperative for efficient tumor staging and follow-up studies. Positron emission tomography (PET) successfully images the lesion metabolic activity. Nonetheless, on course of accurate delineation, chances are high to end up with activity underestimation as, due to the limited resolution, the PET images suffer from partial volume effects. Recently, PET images were modeled as a fuzzy mixture to delineate lesions accurately. We extend this work by proposing a statistical lesion activity computation (SLAC) approach to robustly estimate the total lesion activity (TLA) directly from the modeled partial volume mixtures, without an explicit delineation. To evaluate the proposed method, PET scans of phantoms containing spherical and non-spherical lesions with increased activity uptake were simulated. The PET images were reconstructed with the standard clinically used maximum likelihood expectation maximization and an edge preserving maximum a posteriori (MAP) algorithm, both with resolution recovery. From these images, the TLA was estimated in each lesion using the proposed method and compared to the TLA estimation in the tumor delineations obtained with three state-of-the-art PET delineation schemes. SLAC outperformed TLA estimation via tumor delineation and showed robust against variation in reconstruction parameters. With reference to the ground truth knowledge, SLAC gives median \(\delta \)TLA\(~\approx \) 5 % for spherical lesions. For more realistic non-spherical lesions, median \(\delta \)TLA\(~\approx \) 15 %.
Similar content being viewed by others
References
Weber, W.A., Figlin, R.: Monitoring cancer treatment with PET/CT: does it make a difference? J. Nucl. Med. 48, 36S–44S (2007)
Kelloff, G.J., Hoffman, J.M., Johnson, B., Scher, H.I., Siegel, B.A., Cheng, E.Y., Cheson, B.D., O’Shaughnessy, J., Guyton, K.Z., Mankoff, D.A., Shankar, L., Larson, S.M., Sigman, C.C., Schilsky, R.L., Sullivan, D.C.: Progress and promise of FDG PET imaging for cancer patient management and oncologic drug development. Clin. Cancer Res. 11, 2785–2808 (2005)
Hatt, M., Le Rest, C.C., Aboagye, E.O., Kenny, L.M., Rosso, L., Turkheimer, F.E., Albarghach, N.M., Metges, J.P., Pradier, O., Visvikis, D.: Reproducibility of \(^{18}\) F-FDG and \(3\)’-deoxy-\(3\)’-\(^{18}\)F-fluorothymidine PET tumor volume measurements. J. Nucl. Med. 51, 1368–1376 (2010)
Wahl, R.L., Jacene, H., Kasamon, Y., Lodge, M.A.: From RECIST to PERCIST: evolving considerations for PET response criteria in solid tumors. J. Nucl. Med. 50, 122S–150S (2009)
Chen, D.L., Bedient, T.J., Kozlowski, J., Rosenbluth, D.B., Isakow, W., Ferkol, T.W., Thomas, B., Mintun, M.A., Schuster, D.P., Walter, M.J.: [\(^{18}\) F] Fluorodeoxyglucose positron emission tomography for lung antiinflammatory response evaluation. Am. J. Respir. Crit. Care Med. 180(6), 533–539 (2009)
Lucignani, G., Larson, S.M.: Doctor, What does my future hold? The prognostic values of FDG-PET in solid tumours. Eur. J. Nucl. Med. Mol. Imag. 37, 1032–1038 (2010)
Hatt, M., Visvikis, D., Pradier, O., Le Rest, C.C.: Baseline \(^{18}\) F-FDG PET image-derived parameters for therapy response prediction in oesophageal cancer. Eur. J. Nucl. Med. Mol. Imag. 38(9), 1595–1606 (2011)
Imamura, Y., Azuma, K., Kurata, S., Hattori, S., Sasada, T., Kinoshita, T., Okamoto, M., Kawayama, T., Kaida, H., Ishibashi, M., Aizawa, H.: Prognostic value of SUVmax measurements obtained by FDG-PET in patients with small cell lung cancer receiving chemotherapy. Lung Cancer 71, 49–54 (2011)
Larson, S.M., Erdi, Y., Akhurst, T., Mazumdar, M., Macapinlac, H.A., Finn, R.D., Casillaa, C., Fazzari, M., Srivastavaa, N., Yeung, H.W.D., Humm, J.L., Guillem, J., Downey, R., Karpeh, M., Cohen, A.E., Ginsberg, R.: Tumor treatment response based on visual and quantitative changes in global tumor glycolysis using PET-FDG imaging: the visual response score and the change in total lesion glycolysis. Clin. Positron Imaging 2, 159–171 (1999)
Costelloe, C.M., Macapinlac, H.A., Madewell, J.E., Fitzgerald, N.E., Mawlawi, O.R., Rohren, E.M., Raymond, A.K., Lewis, V.O., Anderson, P.M., Bassett, R.L., Jr., Harrell, R.K., Marom, E.M.: \(^{18}\) F-FDG PET/CT as an indicator of progression-free and overall survival in osteosarcoma. J. Nucl. Med. 50, 340–347 (2009)
Gulec, S.A., Suthar, R.R., Barot, T.C., Pennington, K.: The prognostic value of functional tumor volume and total lesion glycolysis in patients with colorectal cancer liver metastases undergoing \(^{90}\) Y selective internal radiation therapy plus chemotherapy. Eur. J. Nucl. Med. Mol. Imaging 38, 1289–1295 (2011)
Benz, M.R., Allen-Auerbach, M.S., Eilber, F.C., Chen, H.J.J., Dry, S., Phelps, M.E., Czernin, J., Weber, W.A.: Combined assessment of metabolic and volumetric changes for assessment of tumor response in patients with soft-tissue sarcomas. J. Nucl. Med. 49, 1579–1584 (2008)
Francis, R.J., Byrne, M.J., Van der Schaaf, A.A., Boucek, J.A., Nowak, A.K., Phillips, M., Price, R., Patrikeos, A.P., Musk, A.W., Millward, M.J.: Early prediction of response to chemotherapy and survival in malignant pleural mesothelioma using a novel semiautomated 3-dimensional volume-based analysis of serial \(^{18}\) F-FDG PET scans. J. Nucl. Med. 48, 1449–1458 (2007)
George, J., Vunckx, K., Tejpar, S., Deroose, C.M., Nuyts, J., Loeckx, D., Suetens, P.: Quantitative comparison of automated PET volume delineation methodologies using simulated tumor lesions. In: Proc. IEEE ISBI, pp. 653–656 (2011)
Rousset, O., Rahmim, A., Alavi, A., Zaidi, H.: Partial volume correction strategies in PET. PET Clin. 2(2), 235–249 (2007)
Hoetjes, N.J., Van Velden, F.H.P., Hoekstra, O.S., Hoekstra, C.J., Krak, N.C., Lammertsma, A.A., Boellaard, R.: Partial volume correction strategies for quantitative FDG PET in oncology. Eur. J. Nucl. Med. Mol. Imag. 37, 1679–1687 (2010)
Hatt, M., Pogam, A.L., Visvikis, D., Pradier, O., Le Rest, C.C.: Impact of partial-volume effect correction on the predictive and prognostic value of baseline \(^{18}\) F-FDG PET images in esophageal cancer. J. Nucl. Med. 53, 12–20 (2012)
George, J., Vunckx, K., Tejpar, S., Deroose, C.M., Nuyts, J., Loeckx, D., Suetens, P.: Fuzzy statistical unsupervised learning based total lesion metabolic activity estimation in positron emission tomography images. In: Proc. MICCAI MLMI, pp. 233–240 (2011)
Celeux, G., Diebolt, J.: L’algorithme SEM: Un algorithme d’apprentissage probabiliste pour la reconnaissance de mélanges de densités. Revue de statistique appliquée 34(2) (1986)
Caillol, H., Hillion, A., Pieczynski, W.: Fuzzy random fields and unsupervised image segmentation. IEEE Trans. Geosci. Remote Sens. 31, 801–810 (1993)
Caillol, H., Pieczynski, W., Hillion, A.: Estimation of fuzzy Gaussian mixture and unsupervised statistical image segmentation. IEEE Trans. Image Process. 6(3), 425–440 (1997)
Dempster, A.P., Laird, N.M., Jain, D.B.: Maximum likelihood from incomplete data via the EM algorithm. J. R. Stat. Soc. Ser. B 39(1), 1–38 (1977)
Redner, R.A., Walker, H.F.: Mixture densities, maximum likelihood and the EM algorithm. SIAM Rev. 26, 195–239 (1984)
Van Leemput, K., Maes, F., Vandermeulen, D., Suetens, P.: A unifying framework for partial volume segmentation of brain MR images. IEEE Trans. Med. Imag. 22(1), 105–119 (2003)
Hatt, M., Le Rest, C.C., Turzo, A., Roux, C., Visvikis, D.: A fuzzy locally adaptive Bayesian segmentation approach for volume determination in PET. IEEE Trans. Med. Imag. 28(6), 881–893 (2009)
Hatt, M., Le Rest, C.C., Descourt, P., Dekker, A., Ruysscher, D.D., Oellers, M., Lambin, P., Pradier, O., Visvikis, D.: Accurate automatic delineation of heterogeneous functional volumes in positron emission tomography for oncology applications. Int. J. Radiat. Oncol. 77, 301–308 (2010)
Geets, X., Lee, J.A., Bol, A., Lonneux, M., Grégoire, V.: A gradient-based method for segmenting FDG-PET images: methodology and validation. Eur. J. Nucl. Med. Mol. Imag. 34, 1427–1438 (2007)
Van Dalen, J.A., Hoffmann, A.L., Dicken, V., Vogel, W.V., Wiering, B., Ruers, T.J., Karssemeijer, N., Oyen, W.J.G.: A novel iterative method for lesion delineation and volumetric quantification with FDG PET. Nucl. Med. Commun. 28(6), 485–493 (2007)
Dunn, J.C.: A fuzzy relative of the ISODATA process and its use in detecting compact well-separated clusters. J. Cybern. 3(3), 32–57 (1974)
Figueiredo, M.A.T., Jain, A.K.: Unsupervised learning of finite mixture models. IEEE Trans. Patt. Anal. Mach. Intell. 24, 381–396 (2002)
Segars, W.P.: Development of a new dynamic NURBS-based cardiac-torso ( NCAT) phantom. PhD Dissertation, The University of North Carolina (2001)
Reilhac, A., Lartizien, C., Costes, N., Sans, S., Comtat, C., Gunn, R.N., Evans, A.C.: PET-SORTEO: a Monte Carlo-based simulator with high count rate capabilities. IEEE Trans. Nucl. Sci. 51(1), 46–52 (2004)
Hudson, H.M., Larkin, R.S.: Accelerated image reconstruction using ordered subsets of projection data. IEEE Trans. Med. Imag. 13(4), 601–609 (1994)
Bouman, C., Sauer, K.: A generalized Gaussian image model for edge-preserving MAP estimation. IEEE Trans. Image Process. 2(3), 296–310 (1993)
Acknowledgments
This work is financially supported by KU Leuven’s Concerted Research Action GOA/11/006, Research Foundation—Flanders (FWO) and IWT Agency for Innovation by Science and Technology—Applied Biomedical Research (TBM) project 070717.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix A: Parameter estimation
1.1 Gaussian distribution
Let \(\mathcal Y \) be Gaussian distributed \(\mathcal N (\mu ,\sigma ^2)\) with the probability density function \(f_\mathcal{Y }(x) \!=\! \frac{1}{\sqrt{2\pi \sigma ^2}} \exp \left(\!-\frac{(x-\mu )^2}{2\sigma ^2}\!\right)\) and the characteristic function \(\varphi _\mathcal{Y }(\omega ) \!=\! \exp (j\mu \omega -\frac{1}{2}\sigma ^2\omega ^2)\), then from a \(\mathcal N (\mu ,\sigma ^2)\) population with samples \(\mathbf x \!=\!\left\{ x_s \right\} _{s \in \mathcal S }\) where \(\mathcal S \! =\! \{1,\ldots ,N\}\), the maximum likelihood estimates (MLE) are given by
1.2 Laplace distribution
Let \(\mathcal Y \) be Laplace distributed \(\mathcal L (\mu ,\sigma )\) with the probability density function \(f_\mathcal{Y }(x)=\frac{1}{2\sigma } \exp \left(-\frac{|x-\mu |}{\sigma }\right)\) and the characteristic function \(\varphi _\mathcal{Y }(\omega )= \frac{\exp (j\mu \omega )}{1+\sigma ^2\omega ^2}\), then from a \(\mathcal L (\mu ,\sigma )\) population with samples \(\mathbf x \!=\!\left\{ x_s \right\} _{s \in \mathcal S }\) where \(\mathcal S \! =\! \{1,\ldots ,N\}\), the maximum likelihood estimates (MLE) are given by
Appendix B: Convex combination of random variables
1.1 Derivation of probability density function
Let \(\mathcal Y _0\) and \(\mathcal Y _1\) be two random variables with probability density function \(f_\mathcal{Y _0}(x)\) and \(f_\mathcal{Y _1}(y)\), respectively, then their convex combination \(\mathcal Y _s=\alpha \mathcal Y _0+\beta \mathcal Y _1\), will be distributed with density \(f_\mathcal{Y _s}(z)\) as,
Let \(\mathcal Y _0\) and \(\mathcal Y _1\) be independent, then their joint density \(f_\mathcal{Y _0\mathcal Y _1}(x,y)=f_\mathcal{Y _0}(x)f_\mathcal{Y _1}(y)\). Now consider two random variables \(\mathcal Y _0^{\prime }\) and \(\mathcal Y _1^{\prime }\) with densities
Then, (24) becomes
Since the characteristic function \(\varphi _\mathcal{Y _s}(\omega )\) is the Fourier transform of \(f_\mathcal{Y _s}(z)\), from (26), \(\varphi _\mathcal{Y _0^{\prime }}(\omega ), \varphi _\mathcal{Y _1^{\prime }}(\omega )\) and \(\varphi _\mathcal{Y _s}(\omega )\) are related as:
1.2 Gaussian distribution
If \(\mathcal Y _0\) and \(\mathcal Y _1\) are independent and Gaussian distributed with densities \(\mathcal N (\mu _0,\sigma _0^2)\) and \(\mathcal N (\mu _1,\sigma _1^2)\), then from (25), \(\mathcal Y _0^{\prime }\) and \(\mathcal Y _1^{\prime }\) are also Gaussian distributed with densities \(\mathcal N (\alpha \mu _0,\alpha ^2\sigma _0^2)\) and \(\mathcal N (\beta \mu _1,\beta ^2\sigma _1^2)\), respectively. From (27), \(\varphi _\mathcal{Y _s}(\omega )=\exp (j(\alpha \mu _0+\beta \mu _1) \omega -\frac{1}{2}(\alpha ^2\sigma _0^2+\beta ^2\sigma _1^2)\omega ^2)\). Taking the inverse Fourier transform, we can show that \(\mathcal Y _s\) is again Gaussian distributed with density \(\mathcal N (\alpha \mu _0+\beta \mu _1,\alpha ^2\sigma _0^2+\beta ^2\sigma _1^2)\).
1.3 Laplace distribution
If \(\mathcal Y _0\) and \(\mathcal Y _1\) are independent and Laplace distributed with densities \(\mathcal L (\mu _0,\sigma _0)\) and \(\mathcal L (\mu _1,\sigma _1)\), then from (25), \(\mathcal Y _0^{\prime }\) and \(\mathcal Y _1^{\prime }\) are also Laplace distributed with densities \(\mathcal L (\alpha \mu _0,\alpha \sigma _0)\) and \(\mathcal L (\beta \mu _1,\beta \sigma _1)\), respectively. Here \(\mathcal Y _s\) is not Laplace distributed as in the case of Gaussian distribution. To find the density, the inverse Fourier transform can be computed as follows.
From (27), if \(\alpha \sigma _0 \ne \beta \sigma _1\)
Taking the inverse Fourier transform,
If \(\alpha \sigma _0 = \beta \sigma _1 = \sigma \),
Taking the inverse Fourier transform,
Rights and permissions
About this article
Cite this article
George, J., Vunckx, K., Van de Casteele, E. et al. SLAC: Statistical total lesion metabolic activity computation by fuzzy unsupervised learning of PET images. Machine Vision and Applications 24, 1341–1358 (2013). https://doi.org/10.1007/s00138-012-0454-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00138-012-0454-0